This project will demonstrate that study time or injected dose can be reduced by a factor of up to 5 by using time-of-flight (TOF) information with LSO-based PET cameras. This reduction would be achievable in clinical, whole body images obtained with PET cameras that differ little from existing commercial cameras. Therefore, the potential impact of this development is large, especially for oncology. Time-of-flight PET was extensively studied in the 1980's and the improvement in noise performance well documented. TOF research died in the 1990's, as compromises caused by the scintillator materials capable of performing TOF measurements outweighed the TOF advantages. Since then, a new scintillator material (LSO) has been discovered and is now being incorporated into many commercial PET scanners. LSO does not have the disadvantages of previous TOF scintillators, and has achieved 300 ps fwhm timing resolution in laboratory situations. This timing accuracy would easily enable TOF PET. Other factors, such as improvements in photomultiplier tubes and advances in custom integrated circuits, make TOF PET much more practical to achieve today. Therefore, it is very likely that 500 ps timing resolution could be achieved in commercial LSO-based PET cameras, which would reduce the noise variance by a factor of 5 in whole body imaging. Clinically, this variance reduction could reduce image noise, decrease the injected dose, shorten the imaging time, or provide a combination of these three benefits. The tasks in this proposal will demonstrate that the timing resolution in commercial PET scanners can be improved enough to enable TOF. We will first quantify the timing resolution of the components/factors that limit the coincidence timing resolution in commercial LSO-based PET cameras, such as transit time jitter in the photomultiplier tubes, reflections within the scintillator crystals, and performance of the time measurement electronics (constant fraction discriminators and time-to-digital converter circuits). Next we will design, construct, and characterize replacement components for those items that most degrade the timing resolution. We will use these newly developed components to demonstrate, with an eight detector module prototype, the timing resolution that could be achieved in a commercial system. The final task is to quantify the gains that the measured time-of-flight resolution would bring to a modern PET camera performing a whole-body FDG study. The quantification will be done using both 2-D and 3-D TOF reconstruction algorithms and will quantify the improvement in noise and spatial resolution.